https://hanh.me/tags/deep-dive/Recent content in Deep-Dive on Hanh Pham's BlogHugoen-usFri, 23 Oct 2026 09:53:53 +0700https://hanh.me/posts/auditing-karpathys-funnel/Sun, 18 Oct 2026 09:00:00 +0700https://hanh.me/posts/auditing-karpathys-funnel/Every write-up of Karpathy’s four tier funnel repeats his four counts and none of them re-divides them. Against the population that could plausibly use a model, the bottom tier is not 75 per cent of the world, it is 99 per cent of everyone still standing.https://hanh.me/posts/rea-under-the-hood/Thu, 15 Oct 2026 09:00:00 +0700https://hanh.me/posts/rea-under-the-hood/The refusal was more useful than the analysis would have been. Inside REA: how an evidence id is computed so two machines agree, why a zero can mean the question was never asked, and the 3 MB file that exhausted Node’s default heap.https://hanh.me/posts/ios-reversing-mach-o-dyld-shared-cache/Fri, 09 Oct 2026 09:00:00 +0700https://hanh.me/posts/ios-reversing-mach-o-dyld-shared-cache/iOS system libraries ship inside one dyld shared cache instead of on disk, and a fat Mach-O will happily hand you the wrong architecture without an error. Here is the static workflow that reads headers, metadata, and topology before you open a decompiler.https://hanh.me/posts/uncensored-models-are-a-policy-surface/Tue, 06 Oct 2026 15:00:00 +0700https://hanh.me/posts/uncensored-models-are-a-policy-surface/Removing refusal behavior from an open model is technically easy. The hard part is proving what changed, containing the new failure modes, and keeping an agent with tools inside a safe boundary.https://hanh.me/posts/the-oldest-light/Fri, 02 Oct 2026 16:00:00 +0700https://hanh.me/posts/the-oldest-light/We cannot time the universe with a stopwatch. Here is how astrophysicists measure the 13.8-billion-year age of the cosmos using the Cosmic Microwave Background, atomic recombination, and three independent clocks.https://hanh.me/posts/meta-learning-and-meta-cognition-for-engineers/Fri, 02 Oct 2026 14:00:00 +0700https://hanh.me/posts/meta-learning-and-meta-cognition-for-engineers/We will all soon have AI companions integrated into daily life, but passive users risk severe cognitive atrophy. By grounding our workflow in metacognition, the extended mind, and dynamic scaffolding, meta-humans turn personal AI into a high-leverage learning engine.https://hanh.me/posts/the-renaissance-developer/Fri, 02 Oct 2026 10:00:00 +0700https://hanh.me/posts/the-renaissance-developer/When generative AI commoditizes syntax into zero-cost token emissions, the developer’s moat shifts from typing code to systems synthesis: understanding how a single concurrent change ripples across database row locks, cache lines, and cloud bills.https://hanh.me/posts/isa-as-a-contract-risc-vs-cisc/Thu, 01 Oct 2026 10:00:00 +0700https://hanh.me/posts/isa-as-a-contract-risc-vs-cisc/Four decades of micro-op convergence absorbed the classic textbook differences between RISC and CISC, but the instruction set still bills you in three places you cannot optimize away: memory ordering contracts, frontend silicon budgets, and ecosystem sovereignty.https://hanh.me/posts/alan-turing-computable-numbers-entscheidungsproblem/Wed, 30 Sep 2026 16:00:00 +0700https://hanh.me/posts/alan-turing-computable-numbers-entscheidungsproblem/A step-by-step deconstruction of Alan Turing’s 1936 paper: how formalizing the mechanical computer and adapting Cantor’s diagonal argument proved that algorithmic omniscience is mathematically impossible.https://hanh.me/posts/kubernetes-cfs-cpu-throttling-and-resource-limits/Mon, 28 Sep 2026 09:00:00 +0700https://hanh.me/posts/kubernetes-cfs-cpu-throttling-and-resource-limits/Setting hard CPU limits on latency-sensitive microservices destroys p99 response times while average utilization appears deceptively low. Here is how Linux CFS quotas, Go GOMAXPROCS, and GOMEMLIMIT actually behave under load.https://hanh.me/posts/terraform-blast-radius-and-state-decomposition/Thu, 24 Sep 2026 09:00:00 +0700https://hanh.me/posts/terraform-blast-radius-and-state-decomposition/Splitting a monolithic terraform.tfstate into fifty micro-states speeds up plans, but introduces cross-stack read locks, secret leakage, and brittle dependencies. Here is how to structure state boundaries without the integration tax.https://hanh.me/posts/typesafe-jev-decision-models/Tue, 22 Sep 2026 09:00:00 +0700https://hanh.me/posts/typesafe-jev-decision-models/When you force a 70-billion-parameter language model to output ’true’ or ‘false’, you pay for sequential token decoding, uncalibrated probabilities, and broken JSON. Here is why TypeSafe stripped natural language out of the loop.